11.3 � The carbohydrate dietary reference value was previously set on the basis that total fat should contribute about 35% of total dietary energy with protein intake comprising 15% (COMA, 1994). At that time it was recommended that ‘complex carbohydrates’ and sugars in fruits and vegetables should replace some dietary fat, to achieve approximately 50% the proportion of total dietary energy derived from total carbohydrates (COMA, 1994).
11.4 � Concern has been raised that high intakes of total carbohydrate may be deleterious to health, but this report concludes that total carbohydrate intakes, at levels generally recommended in the UK diet, are not associated with the health outcomes examined. Specifically, total carbohydrate intake shows no association with the incidence of cardiovascular disease endpoints, type 2 diabetes mellitus, glycaemia or colo-rectal cancer. In children and adolescents, limited evidence indicates that there is no association between total carbohydrate intake and body mass index or body fatness. Overall, there is no evidence to warrant a change to the total carbohydrate dietary reference value previously set by COMA of approximately 50% total dietary energy.16
11.5 � It is recommended that:
•
The dietary reference value for total carbohydrate should be maintained at a population average of approximately 50% of total dietary energy.17Sugars
11.6 � The previous dietary reference value referred to non-milk extrinsic sugars, and recommended that the population average should be no more than 10% of total dietary energy (COMA, 1991). This dietary reference value was based on the observation that dental caries was rare in populations whose intakes were estimated to be approximately 10% of total dietary energy (based on national sugar supply data from food balance sheets and household consumption data). It was also advised that intakes that exceed 30% of total dietary energy should
16 � The previous recommendation for total carbohydrate was 47% of daily total dietary energy intake or 50% of food energy (excluding alcohol) (COMA, 1991).
be avoided as they may lead to an increase in plasma concentrations of glucose, insulin and lipids.
11.7 � There are several definitions of sugars in current use internationally. The present UK term of ‘non-milk extrinsic sugars’, which is used in national surveys, encompasses sugars in unsweetened fruit juice and honey, as well as sugars that are added to food and drink. It also includes 50% of the weight of sugars found in dried, stewed or canned fruit. The 50% figure is arbitrary and is used to account for the partial breakdown of the cellular structure during processing (Buss et al., 1994; Bates et
al., 2012), but proves problematic when trying to estimate the sugars composition
of certain foods. The term ‘added sugars’ as used by the European Food Safety Authority and US Institute of Medicine includes sugars that are added to foods either during manufacture or by the consumer, but does not capture the sugars present in unsweetened fruit juice or honey. The definition of ‘free sugars’, as used by the World Health Organization, captures all sugars added to foods and those present in fruit juice and honey but does not apply the figure of 50% of sugars in dried and cooked fruit. Therefore, the definition of free sugars is similar to non-milk extrinsic sugars but overcomes the problem of trying to account for the additional sugars from processed fruit. In addition, the term non-milk extrinsic sugars is used exclusively by the UK and does not lend itself to being easily understood compared to the term free sugars or added sugars.
11.8 � In light of the issues above, it is proposed that the UK adopts the definition of ‘free sugars’ in place of ‘non-milk extrinsic sugars’. Free sugars are defined as all monosaccharides and disaccharides added to foods by the manufacturer, cook or consumer, plus sugars naturally present in honey, syrups and unsweetened fruit juices. Under this definition, lactose naturally present in milk and milk products and sugars contained within the cellular structure of foods would be excluded. In this report the term “sugars” is used because this enables other sugars (e.g. glucose, fructose, lactose) to be included in addition to sucrose. This is important because it will ensure that any replacement of sucrose by high fructose corn syrups (an example of a number of different predominantly ‘free’ fructose and glucose mixes with variable but most often 40-60% fructose content) in the production of food and drinks is captured.
11.9 � Since the dietary reference values were last considered, the level of evidence indicating that a high intake of free sugars is detrimental to several health outcomes has strengthened. Higher consumption of sugars, and sugars-containing foods and beverages, is associated with greater risk of dental caries. Randomised controlled trials conducted in adults indicate that increasing or decreasing the percentage of total dietary energy as sugars when consuming an ad libitum diet, either through the substitution of other macronutrient components or by replacing sugars with non-caloric sweeteners, leads to a corresponding increase or decrease in energy intake. Consumption of sugars-sweetened beverages is associated with a greater risk of type 2 diabetes mellitus. Randomised controlled trials conducted in children and adolescents indicate that consumption of sugars-sweetened beverages, as compared with non-calorically sweetened beverages, results in greater weight gain and increases in body mass index; this finding suggests that there is inadequate
energy compensation (degree of voluntary reduction in intake of other foods or drinks), for energy delivered as sugar.
11.10 � The intakes of sugars at the end of the intervention period in trials investigating energy intakes ranged from 4-32% of energy (see paragraphs 6.18 to 6.22). To investigate the relationship between sugar and energy intake further, a meta- regression was conducted, but this found no statistically significant linear dose response relationship between sugars consumption and energy intake (60kJ, 95% CI -0.005 125kJ, change in total dietary energy intake per one percentage change in energy from sugars; p=0.066) (see Figure A9.3 Annex 9). Therefore the precise nature of this relationship remains unknown.
11.11 � To quantify the dietary recommendation for sugars, advice from the Calorie Reduction Expert Group was considered. It was estimated that a 418 kJ/person/ day (100kcal/person/day) reduction in energy intake of the population would address energy imbalance and lead to a moderate degree of weight loss in the majority of individuals (Calorie Reduction Expert Group, 2011). The pooled estimate effect size of 1.01 MJ/day, taken from the updated meta-analysis of trials on sugars and energy intake (see Figure A9.2, Annex 9), was divided by the median between treatment difference in sugars intake of 12.9% (of energy), which results in a 0.078 MJ (78 kJ)/day change in energy intake for each 1 unit change in percentage energy from sugars. To achieve an average reduction in energy intakes of 418 kJ (100kcal/person/day) using this estimated effect size, intake of free sugars would need to be reduced by approximately 5% of total dietary energy (418kJ/78kJ= 5.4). The findings of the trials discussed in paragraphs 6.18-6.21 suggest that there is inadequate energy compensation (degree of voluntary reduction in intake of other foods or drinks), for energy delivered as sugar. The actual, sustained effect size in populations may differ from that observed in these trials, and, therefore, the estimated figure of 5% should be treated with some caution. Nonetheless these calculations provide a useful guide on which to make an informed judgement on the evidence. A 5 percentage point reduction in energy from the current dietary recommendation for sugars would mean that the population average of free sugars should not exceed 5% of total dietary energy.
11.12 � Although the dietary recommendation for free sugars has been derived from calculations assuming that the whole calorie reduction would come from the effects of reduced sugars intakes, it is acknowledged that in reality this would not be the case. High intakes of other macronutrients should probably also be lowered in order for the population’s total dietary energy intake to be reduced. However, lowering sugar intakes provides one approach to lowering the average total dietary energy intake of the population.
11.13 � Given the high rates of obesity in the UK, reducing the population’s energy intakes is likely to be beneficial to health. Obesity occurs when energy intake from food and drink consumption is greater than energy expended over a prolonged period. In the UK, the prevalence of obesity increased sharply during the 1990s and early 2000s. For example in England, the proportion of adults who were categorised as obese (BMI 30kg/m2 or over) increased from 13.2% of men in 1993 to 24.4% in
2012 and from 16.4% of women in 1993 to 25.1% in 2012 (Health and Social Care Information Centre, 2013a). In addition in England, 9.7% of boys and 8.8% of girls (all children 9.3%) aged 4.0-5.0 years and 20.4% of boys and 17.4% of girls (all children 18.9%) aged 10.0-11.0 years were also classified as obese, according to the British 1990 population monitoring definition of obesity (≥95th centile) (Health and Social Care Information Centre, 2013b). Higher rates of obesity were observed in children living in more deprived areas compared to those living in the least deprived. Obesity is associated with greater risks of type 2 diabetes mellitus, hypertension, coronary artery disease and hyperlipidaemia as well as some types of cancer and other diseases (Foresight, 2007). Since free sugars intake is a dietary factor shown to increase energy intake, decreasing the population intake of free sugars is one step that could be taken to help reduce the current UK overconsumption of energy.
11.14 � Reducing consumption of free sugars would also help to lower the risk of dental caries which continues to be a widespread problem in the UK. From the Adult Dental Health Survey, it was found that 31% of adults in England, Wales and Northern Ireland experienced dental caries in either the crown or root of the tooth (Steele & O’Sullivan, 2011). In 2012 almost a third (27.9%) of 5 year olds in England had tooth decay (Public Health England, 2013). There are also stark inequalities across the regions for example, 21.2% of five-year olds had tooth decay in South East England compared to 34.8% in the North West of England with even greater inequalities within local authority areas. In the Children’s Dental Health Survey, 57% of eight year olds had some kind of dental caries in their primary dentition. In terms of permanent dentition, 14%, 34% and 49% of 8, 12 and 15 year olds had obvious dental caries (Lader et al., 2003).
11.15 � In 2013, the prevalence of diabetes mellitus was 6% of the adult population. It is estimated that 90% of these cases are type 2 diabetes mellitus (Diabetes UK, 2014). Given that the evidence in this report found that sugars-sweetened beverages are associated with a higher risk of type 2 diabetes mellitus and that obesity is also linked with this outcome, lowering sugars intake may confer additional health benefits.
11.16 � It is recommended that:
•
The definition for ‘free sugars’ be adopted in the UK and that this comprises all monosaccharides and disaccharides added to foods by the manufacturer, cook or consumer, plus sugars naturally present in honey, syrups and unsweetened fruit juices. Under this definition, lactose naturally present in milk and milk products and sugars contained within the cellular structure of foods would be excluded.•
The population average intake of free sugars should not exceed 5% of total dietary energy for age groups from 2 years upwards.•
The consumption of sugars-sweetened beverages should be minimised, in both children and adults.Dietary fibre
11.17 � In 1991 COMA set a dietary reference value for fibre, defined as non-starch polysaccharides, that recommended the population average intake should be 18g/day, with a minimum of 12 g/day and a maximum of 24 g/day for individuals (COMA, 1991). Non-starch polysaccharides are determined using the Englyst method of analysis (Englyst et al., 1994). Guidance on high intakes of non-starch polysaccharides stated that there was a lack of evidence of benefit associated with intakes in excess of 32g/day. It was noted that such intakes were not seen in self-selected diets and potentially undesirable effects could not be excluded. The dietary reference value was based on the effect of non-starch polysaccharide on increasing faecal weight and the observation that a lower incidence of bowel disease was observed in populations with higher faecal weights.
11.18 � For this report, it was agreed that components would be considered in the context of SACN’s position statement on dietary fibre. This states that for extracted natural carbohydrate components or synthetic carbohydrate products to be defined as dietary fibre, beneficial physiological effects, similar to those demonstrated for the naturally integrated dietary fibre component of foods, must be demonstrated by accepted scientific evidence. Such effects include increasing stool bulk, decreasing intestinal transit time and constipation or the lowering of total cholesterol and LDL-cholesterol concentration (SACN, 2008). However, evidence limited only to effects on gut fermentation or the nature of the microbiota is not sufficient to satisfy this definition. In this report, there is evidence to show that non-digestible oligosaccharides, resistant starch and polydextrose increase faecal mass. On this basis, SACN consider that these three components can be considered as dietary fibre. With the inclusion of non-digestible oligosaccharides, resistant starch and polydextrose this broadens the definition of fibre beyond non-starch polysaccharides. Therefore it is recommended that dietary fibre should be defined as all carbohydrates that are neither digested nor absorbed in the small intestine and have a degree of polymerisation of three or more monomeric units, plus lignin. 11.19 � The majority of the evidence on fibre and health considered in this report uses the AOAC definition of fibre therefore broadening the definition allows the UK to be aligned with the research base and permits intakes to be directly compared with different countries.
11.20 � The broader definition of fibre is measured by AOAC methods and is colloquially known as AOAC fibre. There are different AOAC methods available; the older methods (AOAC 985.29 and 991.43) capture non-starch polysaccharides, some resistant starches, lignin and some inulin, but they do not measure most non- digestible oligosaccharides (Prosky et al., 1988; Lee et al., 1992). It is these methods which have been used in the studies included in this report. A newer method of analysis is now available (AOAC 2009.01) which is able to determine all the components included in the proposed definition of fibre above (McCleary et al., 2010; McCleary et al., 2012).
11.21 � Since the dietary reference values were last considered, the quality of the evidence indicating that a diet rich in dietary fibre (mostly defined as AOAC fibre) reduces the risk of type 2 diabetes mellitus, cardiovascular disease and colo-rectal cancer has strengthened considerably. Despite inconsistency between studies in the definitions of whole grains, greater consumption of whole grains is associated with a lower incidence of cardiovascular disease, hypertension, type 2 diabetes mellitus and colon cancer.
11.22 � Cardiovascular disease, type 2 diabetes mellitus and colo-rectal cancer are of great public health significance in the UK. It is estimated that in 2010, 10.1% of women and 11.7% of men were suffering from cardiovascular disease (coronary heart disease and stroke combined) (British Heart Foundation, 2012) . The prevalence of type 2 diabetes is given in paragraph 11.15. There were approximately 41,600 new cases of colo-rectal cancer diagnosed in the UK in 2011 and it is the third most common cancer in both men and women (Cancer Research UK, 2014a). For deriving the dietary reference value for dietary fibre, non-linear dose-response plots included in meta-analyses of prospective cohort studies investigating cardiovascular disease, coronary heart disease, stroke, type 2 diabetes mellitus and colo-rectal cancer were considered (Figures 11.1-11.5) (Aune et al., 2011; Threapleton et al., 2013b; Threapleton et al., 2013c; Threapleton et al., 2013d). The plots indicate that a non-linear dose-response relationship exists such that as dietary fibre intake increases, the risk of developing these diseases decreases. From these data, it is apparent that intakes of 30g/day and above (as defined using the AOAC methods 985.29 and 991.43) are associated with the greatest health benefits in reducing the incidence of cardiovascular diseases, type 2 diabetes mellitus and colo-rectal cancer. The confidence intervals widen at intakes of 30g/day or more and the health risks are lower at 30g/day than at lower levels of intake. Therefore, it is agreed that the dietary reference value for fibre is set at 30g per day as this is the amount which was shown in the evidence reviewed to be associated most consistently with reduced risk of disease compared to lower intakes. This value was also agreed because the confidence intervals at intakes above 30g/day become much wider and, therefore, there was less certainty about the accuracy of precise values beyond this point.
11.23 � The dietary reference value is based on evidence in which the consumption of a variety of foods rich in dietary fibre as a naturally integrated component is associated with beneficial health outcomes. A diet rich in these foods will also usually be rich in micronutrients and phytochemicals that may have additional health benefits.
11.24 � There is a paucity of evidence in relation to the effects of dietary fibre intake in infants and children upon which to base a dietary reference value, e.g. there are no laxation trials and little information is available with regard to possible adverse effects of high intakes of dietary fibre on growth. The National Diet and Nutrition Survey (NDNS) indicates that the upper 2.5 percentile of non-starch polysaccharide intakes in infants (10-13g/day; 13-17g/day AOAC methods 985.29 and 991.43) and children aged 4-10 (19g/day; 25g/day AOAC methods 985.29 and 991.43) and 11- 18 years (23g/day; 30g/day AOAC methods 985.29 and 991.43) are not associated